Consumer reports on e-book readers

[mogui]

When we are considering the purchase of a new reader we have some tools here to help us make the decision. One valuable one is the matrix https://wiki.mobileread.com/wiki/E-book_Reader_Matrix . Others have written good comments in this section.

Is it possible to create a definitive rating system? Can we establish selection criteria and assign points to each so we can calculate a rating for each device? Here is a first stab at such an approach.

Cost, screen size, battery life and weight are obvious criteria.

Expansion capability via SD cards would be another, as would be USB 2.0 vs 1.1. Included here would be the ability to put new material on the SD card using a card reader so the PC software for the reader would not be needed. Seeing the card as an external USB drive to the PC is a big plus as is the ability to use high-capacity cards.

Would we consider audio capability important enough to consider in our rating system, or is it just a frill?

Screen type. Some people love e-ink while others like backlighting so they can read at night. How should we consider this?

Availability is a consideration. I would not rate Kindle very highly at this writing, but that will change, I hope. Availability of content is crucial. I would like to be able to buy the newest ebooks as well has having access to all of the classics and technical manuals I already have.

Much has been said about reliability in these pages. How can we quantify this?

Ease of use is another consideration. Maybe we could do a poll on this.

Does the reader include a basic feature set such as bookmarking, page rotation, wireless connectivity, ease of holding, outdoor readability, user maintainability (e.g. battery replacement), openness vs DRMness and competition among available ebook vendors, fonts.

Format compatibility is a big issue. I think a good reader would host many file types natively, but that is only my opinion. The two most common ebook formats for technical books are PDF and HTML. Legacy ebooks for the Palm are often in PDB format. The Microsoft Reader format (LIT) is also common, as is TXT. Microsoft Word compatibility is also useful (DOC, RTF). Some people would find image display capability (JPG) important. How can we assign a value to file compatibility? Is it enough that the PC or Mac software does the conversion? Or should the device do the conversion? How many formats are enough? Are good authoring tools avalable? Are the ebooks you buy for the device usable on another device?

Are there more criteria to consider? Is there a convenient way to calculate a score? Can we generate a poll for issues such as reliability and ergonomics? Can we identify factors which should be subjective considerations vs absolute criteria?

These issues are complicated for the buyer to contemplate. Is there a mathematical method we could use to pare the decision down to one of simply evaluating the subjective criteria and using a rating system for the rest of the puschase decision? This is perhaps an opportunity to give the ebook designers a way to evaluate their new designs.

[Alexander Turcic]

This is intriguing. It would be cool if we could come up with some formulas.

And it sounds like you already have practice evaluting products in quantitative terms.

[mogui]

Originally Posted by Alexander Turcic

. . .sounds like you already have practice evaluating products in quantitative terms.

Not so much Alexander, but maybe this could be started in a spreadsheet. It would be easy to adjust the weighting of the scores to achieve a rational conclusion.

I looked at Google Spreadsheets to see if we could collaborate online, but everyone would need to send their email address to a central coordinator to gain access. It sounds cumbersome. It would be nice to have an embedded spreadsheet right on this page. Is there a spreadsheet wiki? Any ideas?

[nekokami]

I don't think you can come up with a single rating for each product, as different people weight the features differently. For example, stylus input is quite important to me, but to others, much less so. But your idea of a spreadsheet where a potential user could weight several factors and then get ratings of the different units would be good. You've got a nice list of requirements to start with. I'd add the OS on the unit -- I strongly prefer Linux-based models, myself. There are also folks who want to be able to interface with the unit using Linux, rather than a PC or Mac. That could be another factor.

Tiki has spreadsheets built-in. I have a couple of Tiki installations and could possibly donate space, if that would be helpful. Tiki also has polls and surveys available, so you could let users rate the reliability or ergonomics of various units, trusting that people won't rate units they don't own.

[BACbKA]

Host OS supported (actually as well as by manufacturer) is an important factor IMHO.

[mogui]

Originally Posted by nekokami

Tiki has spreadsheets built-in. I have a couple of Tiki installations and could possibly donate space, if that would be helpful. Tiki also has polls and surveys available, so you could let users rate the reliability or ergonomics of various units, trusting that people won't rate units they don't own.

That sounds great Neko! I really need to get up to speed on wikis other than tiddlywiki and this would help. Can we have a spreadsheet that is editable by any visitor? Polls and surveys would be a big help too.

Does anybody know if the mobileread wiki has this capability already?

I suggest we start by trying to define absolute criteria wherein there is no debate over the objective nature of each criterion. This could be used to develop the basic score. This would be useful by itself. The remaining criteria could be the subjective ones -- personal preferences. Thus we would have two sets of criteria. The matrix would display a pre-calculated objective score for each reader.

The user would then be able to generate a total score relative to his/her personal preferences by entering data into the matrix for those criteria that reflect their personal needs. Then the numbers could be crunched over a range of possible readers calculating the subjective score and adding it to the absolute score for a total rating customized for that user.

Maybe it is possible to have poll results feed cells in the spreadsheet.

[mogui]

Neko was kind enough to give me a page on her wiki to develop the spreadsheet for the reader evaluation matrix. However, the only spreadsheet I have used, ever, is the one on my Palm Pilot and I have to read the manual first to do anything -- every time. So I decided that if I am to break something, it is better that I break Google than Neko's wiki. Therefore I opened a Google Docs account and made it publically viewable. I think the spreadsheet is a good approach, but I think I will need help. I have worked for a while to get the title on this spreadsheet and all I have accomplished is a headache.

You can edit it if you like. Send me a PM and I will add your email address to the permissions list on the site. I will work on this as my tolerance for frustration permits, but it may be slow.

My intent is to take criteria from the top post in this thread and place them in the first column. The top section would be criteria that are deemed objective -- that is, important to everyone and not arguable. The bottom half would contain subjective criteria. The user would rate features in this section according to desireability. So the second column would contain the weighting factor for that particular criterion which would be used in calculating the score for each reader. The spreadsheet can be exported so anyone can adapt it to a special purpose.

The top row shoud be the title. The second row would contain column headers for each reader type.

As you can see, I am stuck on entering the title.

Update: I managed to get row and column headers in place. headache is gone. I have not attempted any formulae. Please feel free to add or change whatever you like.
Update: The UI for the Google Spreadsheet is quite nice. There is plenty of help on the site. I find I can figure out most things. I do not have any formulae in the sheet yet, so updating is still quite fast. This utility has versioning, so I don't have to worry too much about making a mistake. Follow the link here to view the reader evaluation matrix. Send me a PM so you can get added to the list of permitted modifiers.