[globular-toast]

I think it's a case of damned if you do, damned if you don't. If releases weren't frequent people would have to build their own from the master branch. A system package manager makes updates really low friction. Ubuntu seems to be only a few months behind the upstream releases. IMO individual packages should not be reinventing the package manager wheel.

The main problem with Calibre is it's quite hard to package and many distros seem to have given up.

[KennyBlanken]

The releases are frequent in part because Goyal refuses to separate all the content-based stuff (parsers for online news sources, for example) from the application.

This is an extra pain in the ass for users because if your favorite news site parser breaks, you have no choice but to upgrade Calibre itself instead of just updating a config file.

At a minimum the news-parsing stuff should have long ago been split out into a plugin. It has nothing to do with calibre's function as an ebook library tool.

[aumerle]

The updates have nothing to do with news parsers. News parsers are loaded dynamically, every time they are used. So are metadata downloading plugins. So are get books stores. Indeed almost all code that parses data from the web in calibre is loaded dynamically independent of calibre updates. The calibre developer whom you so blithely complain about here actually jumps through a million hoops to make sure those bits of calibre remain backward compatible with version of calibre and therefore python, going back a decade and more.

[xbmcuser]

I agree with you but I also think that calibre should have update patches instead of downloading and installing 100mb file every 2 weeks. Calibre comes with its own python it is mostly self contained so it would be great if we had a choice for downloading just an update patch. I don't think this is something Kovid has the time though he probably could do it. It's the same problem when he was not willing earlier to move to python 3 as the amount of man hours required vs maintaining Python2 for calibre himself was easier. But someone came along and started porting calibre to python 3 and Kovid used their help to move to python 3. So if a dev comes along with better way to update calibre Kovid will probably accept it but as it is not a priority for him he might not ever get around to doing it himself.

[alsodumb]

This is something I often tell my friends who complain about Calibre: his project his rules. If you think there are genuine pain points you can address, fork it and make a better one no one is stopping you.

[telmo]

> This is something I often tell my friends who complain about Calibre: his project his rules.

It is hard to imagine that anyone would disagree with this. I certainly don't, nor did I say anything to the contrary.

> If you think there are genuine pain points you can address, fork it and make a better one no one is stopping you.

The only thing stopping me is common sense.

[telmo]

> If releases weren't frequent people would have to build their own from the master branch.

But why??? What in the world of Calibre demands such urgent updates? I use a variety of software (some quite complex) that do a great job with much less frequent updates. Why not Calibre?

[caslon]

A big portion of what people use Calibre for is parsing, converting, and generating HTML. The epub file format, in practice, is zipped, poorly-specified and poorly-generated HTML. It is far from an exact science, and minor improvements to this can result in countless lives being saved slight amounts of discomfort from terrible HTML-handling or poorly-handled OCR.

To give an example, I once needed to reference a 2003 book that was out of print for a long-running project I was working on at the time, and the publisher who released it digitally had gone out of business, and the author's mailserver had gone dark. I checked the usual suspects for the PDF or hard copy of the book, but the only remaining copy of it I could find was a Calibre automated epub conversion from quite a few years in the past. A few years later, I once again looked for a better copy, only to find another Calibre epub conversion, but still no PDF in sight. This one, seemingly generated a few years later, was much higher in quality.

I was pleasantly surprised.