I'm sure calibre 2.0 is a great technical feat, and kudos for all the work put into this product, but judging by the screenshot the user interface is equally clunky as in 1.0.
This software does mostly everything I need it for to convert ebooks... but can you, the authors, please improve the UI?
I was so excited to see this headline specifically because I hoped they would be doing a UI overhaul. Unfortunately, that just seems to not be the case or even on the horizon. I'd definitely consider dedicating time with it, but I've heard nothing but bad things about trying to work with the maintainers of the code.
Calibre makes it dead simple to manage a library of ebooks and allows you to do just about anything that can be done with them. That's the definition of a good UI. Whether it follows modern design trends is irrelevant. A UI overhaul would be a terrible idea.
Those are very undefined and user-specific complaints you have. And I've seen nothing in your post that justifies a full UI rewrite. Perhaps some changes, to cater for new/novice/confused users, but a UI rewrite is just a ridiculous thing to request.
Considering that it's free and open source, and that I use Calibre for a fraction of a percent of the time that I do actually reading the books on my Kindle that I loaded using Calibre, I would rather that the creator of Calibre focused on the technical details, such as making sure that the file conversion/transmission is working correctly.
While that's absolutely desirable in terms of priority, calibre's UI has been not merely ugly for its entire six-year history, it's been a jet-powered unicycle ride to Crazytown for its entire six-year history. We're talking about things like just following generally accepted conventions for how to structure GUI menus.
I don't think this is a matter of calibre de-prioritizing fixing the UI. I'm pretty sure it's a conscious choice. They know their UI backward and forward and it works for them, therefore anyone who complains about the UI Just Doesn't Get It. That unicycle is BEAUTIFUL, dammit.
I'd like to see a few more jet-powered unicycles rides. They sound fun. Functionality, especially in niche software, is sometimes more important than some kind of idealised smooth workflow.
The nearest physical analogue I can conjure up is to contrast a 'sound system' as used in dances in church halls and community centres round here (wardrobe sized speaker cabinets with 18" paper cone speakers, horn tweeters, all patched into a rack based preamplifier that only the constructor and a few close associates know how to navigate) and a domestic hifi by (say) Denon. Both have their place.
I don't disagree with your position -- I disagree with the implicit contention that calibre's non-standard UI improves its functionality. It is, at its heart, a file management and viewing application. What advantage does it get by spurning the notion of File and View menus? How is it made better by having a toolbar full of non-standard icons that literally duplicate the menu bar? Is it merely UX idealism that would lead one to argue that menus that have items twenty words long is needlessly suboptimal?
There are GUI conventions which at this point have been with us for three decades, and no matter what operating system you're using the vast majority of programs you use follow them. I would never argue that it's impossible for a program to be improved by deliberately flouting those conventions -- but I'd argue that you need a defensible rationale for doing so. A sound system for a large space has different tradeoffs and design considerations than a home theatre system does, but it's actually following a different set of well-considered conventions, right? I don't think calibre's UX design is well-considered at all; I think it was put together by someone who doesn't know much about UX design, isn't willing to learn, and whose reaction to even gentle criticism is to dig in his heels.
>Functionality, especially in niche software, is sometimes more important than some kind of idealised smooth workflow.
I buy this argument for professional GIS software, database management software, etc. But come one, Calibre basically has 3 purposes: keep a library of ebooks, convert them between formats, transfer to devices. It shouldn't be so complicated. Every time I go open it I have to relearn how it works.
> Every time I go open it I have to relearn how it works.
Then you must be doing something wrong. I use Calibre very infrequently, and yet I've never had any trouble remembering how to convert a book and load it onto my Kindle.
For file conversions, I find myself relying almost entirely on pandoc. Commandline tools rock.
Calibre has done a few conversions, but given the typical use-case I've found is for trying to get PDFs into anything else vaguely usable, it's almost always a complete wash.
Going to Markdown or LaTeX and from that to the format(s) I want is almost always the preferable route.
I use calibre for converting web pages I want to read offline into epub, but then I use the file manager to transfer the epub onto my device - it seems to be the easiest way.
Also, on every single conversion I have to tell it again that I don't want a huge margin. Why would I want to make my e-reader's screen ever smaller?
Maybe it's best to build a whole new app out of the low-level libraries and tools that make calibre. A simple app with a simple UI.
I've seen this complaint with lots of Open Source programs and, while your opinion is valid, I also can't help but feel that you are being a bit ungrateful.
I disagree. This is one of the most important pieces of feedback any opensource project can get. Technical merit means nothing if users find the cost of using the interface more trouble than the benefit of the program.
I think the lack of emphasis on creating polished user interfaces is a real ideological stumbling point for the open source movement.
Given that the thread you're responding to leads in with the suggestion to contribute code to improve the application, your point is both factual and totally tangental to the conversation.
If the project is hostile to merging patches (which Calibre's developer had demonstrated), that is relevant when discussing the possibility of a user providing contributions for improving the UI.
