I Would Rather Use a Kindle Than a Kobo
Contents
I once owned a jailbroken Kindle Paperwhite 12th Generation.
I had a couple of plugins installed, and it was forever locked to its preinstalled firmware (5.17.10, I believe). But I was beginning to think that the overall Kindle experience, regardless of whether it was jailbroken or not, was inferior to what I had heard about other e-reader devices, specifically the Kobo devices. So I soon became interested in trying a different device.
According to users, reading on a Kobo was a more liberating experience than reading on a Kindle.
Users claimed that the experience was much better simply because it could natively read .epub files (although people usually recommend kepub instead of epub, and never mention this beforehand), because you can customize various parts of the user interface to better suit your preferences, because you can install KOReader without having to jailbreak the device first, but, above all, because of its freer, unrestricted ecosystem. For those users, it was clear which of the two e-readers was the better choice.
And, after a while, I returned my Kindle and bought a brand new Kobo Libra Color. I had my own reasons for why, at the time, I considered the drawbacks of owning a Kindle to be serious enough to justify switching brands.
But once the “honeymoon” with this new device was over (a honey-noon of sorts) I gradually began to feel that perhaps a Kindle was a much better option than the Kobo.
Why did I switch in the first place?
Death by Thousand Papercuts Broken Updates
Starting in 2021, Kindle updates became rather annoying to deal with because they would often break some functionality of the simple device that reads books. The manga UI would stop working, the virtual panel would break, and sometimes even basic features such as the dictionary or the ability to read sideloaded PDFs would stop functioning properly. There's always some feature in each update that gets broken for either a week or a month until it is finally fixed, and once it gets hotfixed, something else would break, repeating the cycle.
As if that weren't enough, the end-of-life updates released for older Kindles would also end up making the device feel sluggish, which a lot of users theorize is done on purpose to encourage people to buy a brand new Kindle.
A few months ago, people were worried that their Paperwhite 11th Generation (the one released in 2021) had become End-of-Life because of the lack of updates while other models were getting the hotfixes just days after the bugs were reported. This still seems to be the case, because this particular model hasn't received a single update since April 2026 (this article is currently being written on August 8th).
Just around the time before I switched to Kobo, there were a lot of supply-chain attacks targeting various technologies, and there were also rumors that the newest upcoming jailbreak for Kindle was so dangerous that the developers behind it had to contact Amazon before releasing it (until very recently I learned that Amazon never let them release it). There have been a few exploits on Kindle that could actually make your device, and possibly your Amazon account, vulnerable, with KindleDrip being the biggest of them all.
By simply knowing the email address assigned to your device, an attacker could send a book to your Kindle cloud library, and if you clicked on a link inside that book, it would load the browser and a maliciously modified JPEG XR image. This could exploit a privilege escalation vulnerability and, if the attack was sophisticated enough, it would append a credential-stealing script to the first payload. The exploit was fortunately reported before anything global could happen, and although Amazon took its precious time to act on it, it was successfully patched months later.
A few articles came out of this, along with observations about how easy the exploit was to execute remotely and even a brand new jailbreak method based on it, but probably not much else. It was 2021. Amazon and developers around the world were still doing their jobs to safeguard their software and trying to keep everything relatively safe.
But now it's 2026, and everyone and their dog is using AI to print "Hello World" on top of using three JavaScript frameworks and four abstraction layers, creating amounts of technical debt unlike anything we've seen before.
They are more free to use AI to do so, I can't make developers not use AI for anything they wanted to code, my concern here comes from the fact that they have added the instability and volatility of AI on top of their already fractured, broken and rushed work pipeline.
Poorly written pieces of software that simply existed and didn't bothering anyone are now being patched, refactored, and extended by tools that produce code faster than anyone can properly understand or review it. This of course, never ends up in necessarily better software, it simply means that instability is almost invisibly being reproduced at a much greater speed.
And software developers aren't the only ones with access to these tools. While one group is using AI to fix spelling errors on code, another is using it to inspect the entire repository, looking for anything that can be exploited. The people writing the software may be moving faster now, but so are the people looking for ways to break it.
There are thousands of users running AI agents in anything that can execute code, trying to uncover the slightest vulnerability possible just so they can disclose it on GitHub and Hackernews before telling the developers for clout. npm and various other similar ecosystems, for example, are currently being torn apart by endless supply-chain attacks, and there don't seem to be any concrete ways to protect against them beyond advising users not to install or run unknown packages or programs, without first acknowledging that their entire infrastructure is built on this very concept of trust.
The current software ecosystem, as we know it today, is being stunted by endless bureaucratic Trello SCRUM pipelines, abstract layers and as of lately, developer's inability to write and review code made by AI.
It's more than likely that the operating system on Amazon's Kindle is going through the same thing. On top of the problems mentioned above, we now have to consider the possibility that some of the increasingly slow, bug-ridden final and even latest updates are the result of excessive "help" from artificial intelligence and rushed patches being pushed into production.
Gone are the days when we merely suspected that the final updates for older models were slowing down the device to encourage customers to buy the newest model.
For current models, you would need to wait for the newest patch, for older and unsupported devices, you would be stuck with that broken firmware update forever.
At the time, I put myself in the position of having to pick one of the following:
Would I keep my device unjailbroken and updated to the latest firmware to keep my Kindle as safe as possible, in exchange for having one or two broken features? Or would I block the updates, stay on an older firmware version, jailbreak it, and hope that another global KindleDrip exploit or a vulnerable extension update wouldn't come along?
The answer for many readers would be the latter, because KindleDrip-level exploits are rare, and the plugin ecosystem, paired with KOReader and its greater popularity, makes it the better option overall.
For them.
Putting aside the software concerns mentioned earlier, even if I stayed on the latest software, I wouldn't be able to bear the broken features that came with it, as they were the reason I initially switched to Kobo.
I didn't wanted to update my Kindle only to find out that it was slower than before, or that one or two features were broken and wouldn't be fixed until the end of the month. Nor did I wanted a firmware-locked and jailbroken Kindle that could potentially be hit by a vulnerable extension update because of a developer account was compromised by a supply-chain attack somewhere downstream.
I also didn't care that much about KOReader, even if users suggested it would "enhance" the reading experience.
I am more than sure that another KindleDrip-level exploit won't happen anytime soon, but before I switched, it seemed like every piece of infrastructure was bound to be found vulnerable sooner or later, which made me more cautious about keeping a jailbroken device as my primary e-reader. And considering that every piece of software nowadays is blindly aided by AI and, as a result, becomes a little bit unstable or even unsafe, it seemed to me that anything running on my jailbroken Kindle could also eventually be affected by anything.
Adding to that the fact that Kindle was going to block access to its account login services on some older models, along with the videos I had seen showing off Kobo's freer ecosystem, I decided to pull the plug, return the Kindle, and buy a Kobo.
Although, very surprisingly, this is where I started noticing cracks on the other side of the walled garden.
Buying a Model
For starters, Rakuten does not currently offer a 7 inch black and white model online or in physical stores. They only offer a Clara B&W model in 6 inch online, while every local store I've visited only offers the Clara and Libra in color, which means that this particular B&W model seems to exist only online.
Most of the older B&W models have been removed from the shelves, both online and in physical stores, and it is such a disappointment because I was interested in getting any of the older models, such as the Sage, Elipsa, Forma, or Libra 2, which was also a popular option among users.
I didn't wanted to downgrade from the 7 inch screen of my 12th generation Paperwhite to a mere 6-inch Clara, especially because it had a recessed screen.
In fact, almost all Kobo models have recessed screens including the "pro" versions, which is something a lot of people like to criticize when talking about Kindle models, yet you don't see nearly as many people talking about it when someone switches to a Kobo.
I know I could get one of the older models from eBay or even AliExpress, but a quick search on either platform will tell you why I ultimately decided against it. The models you are most likely to find are often the "basic" ones from before 2018, while the slightly newer ones are listed at the same price as the current models.
And, as I explained in my iPod De-influence article, shipping will push that price even higher. Just to give you an example, a Kobo Glo N613, with a 6 inch screen and 2GB of storage from 2012, is listed at $46 on eBay, but the price jumps to $120 once shipping is added.
And it will arrive in 2 months.
I don't live that far away from the United States, and yet it seems that I am being priced out just because there are two borders in between.
My luck in finding a second-hand Kobo would probably be better if I searched on AliExpress, but that luck only holds if I'm looking for an older model such as the Kobo Glo HD or the H2O. Any current or preowned B&W 7 inch model will be listed at roughly the same price as when it was originally sold.
In fact, I am currently finding a Kobo Libra 2 on AliExpress, and the only listings I could find are selling it for $370.
Great.
Stubborn in geting a 7 inch Kobo, I decided to visit my local store and buy the latest Kobo Libra Color. It set me back $300, which is a lot more expensive than the Kindle Paperwhite 12th Generation, quite a lot more expensive actually, but at least the specifications of this new model are miles beyond what any of the current Kindle lineup offers, a color screen, 32GB of storage (with expandable storage if you know how to crack open the case), stylus support, buttons (something I am not too crazy about), and, although quite expensive again because of shipping, a full replacement kit from iFixit.
You might also wonder, why didn't I buy an e-reader from another competitor?
Well, you see, e-reader technology is in fact expensive to manufacture compared to standard LCD or LED screens. We are talking about a technology based on ink that is not only a niche on its own, but turning it into an e-reader makes it even more niche, which ultimately allows brands to charge whatever they want for it.
Amazon is the only brand that can afford to sell the Kindle at an affordable price because they are technically selling them at a loss, expecting to recoup that loss through Kindle Unlimited subscriptions, which, if you haven't tried it, is actually pretty nice to have.
If we wanted to talk more about prices, a quick search online will also show you that most 6 inch B&W models from other brands are in the range of $150 or more. Smaller models, similar in size to smartphones, are in the $100 range. They are interesting and affordable devices if you want to read everything on a screen the size of a smartphone. I personally would rather stab my eyes than do so, but some people find them approachable.
Pocketbook, another strong competitor, offers basic models at $175 which is a decent price but unfortunately, most of their "basic" are still stuck in 6 inch screens, even then their 7 or 8 inch screens seem to like sitting in the $300 range dollars for some reason.
Bigme e-readers are also available at the price of $400 dollars, but I believe this price comes from the fact that these tablets are more interested in being used for drawing and note-taking on a larger screen (they're mostly offered in 8 inches or more).
Same thing with Remarkable and Oonyx Book, they have larger screens, but their interest comes in drawing and note-taking. They sit at $468 dollars.
There's also the fact that most other e-readers have Android running on them, which in my opinion is not something I am looking for.
I actually wanted something that acted as a reading device, that had a 7 inch screen, and not something that has an unoptimized Android instance and the cheapest kind of e-ink based screen slapped into it.
I did my research, I analyzed my preferences, watched some reviews and came to the conclusion that a Kobo Libra Color suited me best, despite its unnecessary color screen, recessed screen and larger price comparison to a Kindle.
Some time after making my purchase, I found out that Rakuten had increased its price twice for most of their models. I made the purchase on a physical store because Amazon was, somehow, offering the Libra Color at an even higher price online (I also did it on a physical one because I wanted to support the local store).
September 13th Update: Amazon also increased Kindle's prices in late August, and they remain cheaper than Kobo's!
This is another important consideration when comparing e-reader prices around the world, not everyone gets access to the same prices that these brands offer in the U.S, Canada, or the United Kingdom. A lot of people criticize others for buying a Kindle, even after pointing out the damage Amazon has done to the book industry and the privacy concerns surrounding its data-harvesting practices, which, in one firmware update, were responsible for mistakenly uploading GBs of data and draining battery life.
What people don't mention in their criticisms is the fact that Kindles are, on top of being the most popular e-reader brand, one of the only brands that sells e-readers at an affordable price, again because they sell them at a loss.
In other countries from my continent, Kindle devices might even be sold for cheaper, while competitors such as Kobo, Onyx, and PocketBook can only offer them at a fixed price because they aren't as large as Amazon.
I do have the luxury of buying an expensive and feature-rich device, but what would be the case for everyone else not living in the usual countries?
After comparing prices, hunting for older models, visiting stores, and convincing myself that the extra cost was justified by the features, I finally had my Kobo. There was nothing left to do but use it, would I finally enjoy the supposed liberating experience of the Kobo?
Death by Thousand Lacerations
I want you, the reader, to understand that I am fully aware that some of the following issues may not affect you, or may have never happened to you, and I am happy for you if that's the case. Unfortunately, you are not the main character on this big planet called Earth, and neither are your experiences or opinions universal. "This issue doesn't happen to me". I don't care. It has happened to me and other people.
Kobo, somehow, manages to fail at the things I care about most in an e-reader.
Starting with its database management, it's built in such a horrendous way that it almost makes Kindle look good in comparison.
For those unaware, Kobo can randomly corrupt its database. Users don't know why this happens, developers are ignorant to this issue, and the fact that it can still happen at all, on its own, makes it such a huge problem that makes it impossible to ignore it without taking precautions. Your Kobo's database handles your annotations, reading progress, book index, collections, and user settings. If your Kobo decides it has had enough, all of this can be lost the next time you connect it to your computer.
The band-aid solution users suggest is to back up your entire .kobo folder, either manually or with the help of a Calibre plugin, and restore it if the database becomes corrupted.
If you come to Reddit or MobileRead asking for help, you will probably get the response that you should have backed up your database and that it was all your fault. This suggestion is, of course, understandable, you should back up everything you can. But this suggestion falls short because the same user didn't even know this was a problem in the first place.
Some users suggest that, because Calibre reads and modifies the database, it could be responsible for the corruption. But I would argue that this may have more to do with users failing to safely eject their Kobo from their computer, leaving the database with missing or incomplete information that can easily lead to corruption.
Sometimes, your Kobo can display an error indicating that the database is corrupted, but oddly enough, a simple restart fixes it. In fact, many of the bugs that appear after connecting a Kobo to a computer, such as being unable to read metadata from the device, edit configuration files, or upload files, can be solved simply by rebooting the device.
This would be a relief if it didn't happen so frequently.
Thus, one of the more reliable ways to prevent this problem is to install a third-party plugin named Kobo Utilities, click on it while browsing your device through Calibre and select the "Backup database" option.
Amazing.
At least with Kobo, you can sideload books seamlessly through a simple USB transfer, Google Drive or Dropbox integration, or even through the "experimental" browser! All of these are great options, except for the last one, as it will, just like Calibre does by default, store the files in the root directory.
If you are an organized person, it will be unbearable to see 70+ stray book files sitting in the root directory beside the default Kobo folders, or, in the case of Calibre, thousands of folders named after authors' last names.
For those who don't mind this, you would still need to know that the database will permanently associate that book with its original indexed location.
This means that if you read or annotate anything on the book and later decide to move it somewhere else, such as the humble and custom /mybooks folder, the database will no longer recognize the book's new location and will reset the statistics for that book. It will treat it as if you never had the book in the first place.
I would guess there is a way to restore the settings for that book by modifying something in the SQL database, but I wouldn't count on it.
At least the books transferred through the official Google Drive and Dropbox integrations will be happily stored within the .kobo/ folder without any complaints.
For those wondering, these integrations almost work like the Send to Kindle tool. You install Google Drive or Dropbox on your smartphone, and I would guess that, by clicking the Share button on an .epub file, it will show you a quick sharing option for the Rakuten folder. It's not email, it does not sync progress, but it works almost the same way if you have also used the official Kindle app.
At one point, I tried adding an entry to NickelMenu to automatically move every .epub file found in the root directory into a custom folder. I couldn't get it to work, unfortunately, and decided to stick with Calibre.
See, I admire how Calibre has stood the test of time, but I don't like how it handles a few things regarding device connections and book collections. Because on top of doing a lot of things its own way, it also seems to believe it has the luxury to fail on you.
It has a lot of features, yes, but you also have to spend a few hours with it just to diagnose and troubleshoot them, let alone customize them to your liking.
Kobo Utilities is among the most recommended external plugins for Calibre. It offers an option to keep multiple backups of your database, successfully updates metadata from books modified in Calibre (because the default tool likes to not work), removes extra covers from deleted books, and will try, and I really do mean try, to back up and store your reading progress, rating, time spent reading, bookmarks, and reading position in Calibre.
I really mean try, because it has never worked in any of my four attempts.
Not only do I now have four extra columns on my home screen, but my precious original epub files have also been contaminated with random and empty Calibre entries in their metadata, with nothing in them because either Kobo Utilities or Calibre didn't want to work as they were supposed to. Right I have to make a new folder that only includes my original epubs.
The only tools from Kobo Utilities I use nowadays are the ones for cleaning up extra covers and automatically backing up the database, nothing else.
After all, I don't need the extra fluff from most plugins found on Calibre or in the Kobo Developer's Corner. I only need to store my books, read them, and export the annotations, nothing else.
Except that Kobo fails at one of those tasks.
Annotating.
One of the greatest and advantageous feature of e-readers over physical books, is the ability to annotate, write notes and export them to a simple file.
Kobo does these three things horribly.
For starters, annotating a paragraph does not automatically snap to either the beginning of the paragraph or the period at the end. It's a free for all. If the paragraph ends halfway through a line, Kobo will snap to the end of that line instead of the period, leaving empty space after exporting the highlight. Even then, it will sometimes snap to a symbol, either a comma or an em dash, but only in those moments when you don't want those symbols included in the annotation.
To make matters worse, it will not preserve line breaks in bulleted paragraphs, even on their precious .kepub file format.
So you can expect your exports to look like this:
The behavior of a system cannot be known just by knowing the elements of which the system is made.
^
[end of line]
THINK ABOUT THISHow to know whether you are looking at a system or just a bunch of stuff:A ) Can you identify parts? . . . andB) Do the parts affect each other? . . . andC) Do the parts together produce an effect that is different from the effect of each part on its own? . . . and perhapsD) Does the effect, the behavior over time, persist in a variety of circumstances?
Yes, I have tried annotating with other books, they all vary in success. Both epub and kepub files, none of them good enough to be usable after exporting, don't try arguing with me about this.
I don't understand this. Kindle's and KOReader's annotating tools work so much better than this. KOReader can be a little stubborn about selecting the period sometimes, but it is far less troublesome when annotating across pages. Kindle can also be a little sensitive when annotating, but that's understandable because it is the fastest of the three at selecting text. Kobo offers nothing in return but endless friction.
It also does not offer the option to export annotations from sideloaded books by default. You have to add a line (ExportHighlights = true) to the Kobo configuration file, reboot the device, and click the newly added "Export Annotations" button. You would think that would be it. But no, there's something worse than all of this.
It does not include chapters.
It also does not include color annotations.
The second most important thing to me when it comes to e-readers is annotations. The first is that the annotation export includes chapters.
Kindle semi-does this. KOReader does this through its 10+ export options. Why doesn't Kobo? The fact that Kobo doesn't include color annotations in its export files is just rubbing salt in the wound at this point.
Exporting the annotations through Calibre, or rather specific, through the Kobo Utilities plugin, will include the chapters, but I am writing this experience in the context of someone who wouldn't use Calibre and is solely trying to export their annotations natively.
There are a couple of services that manage to sort of do this in their own way. Searching on the Internet, you will find several free Kobo Highlight Exporter tools, but they all vary in how successful they are, some produce duplicated annotations, some include colors but miss chapters, others include subchapters but also miss the colors and fail to properly index other annotated paragraphs, and so on.
Even if you find one website that works for you, it will probably go down after a year, without instructions of how to host the page yourself.
When I realized this during my second week of using Kobo, I knew I had to use (and even expected to pay for) a good exporter tool, and thus, landed on Readwise, it is the most popular and advertised service out there. It has plenty of export options and its dashboard seemed quite complete.
I used the 30 day free trial and found out that I had to manually load my Kobo database in order to export my highlighted paragraphs. I even had to download a separate program to do so. The setup wasn't complicated and I was eventually able to see my notes on the website.
But still, it didn't include chapters or even the colored annotations.
It may be more convenient to export notes from purchased books rather than sideloaded ones, but this still feels, to some degree, like a second-class experience compared to how Kindle handles annotations on uploaded books or even sideloaded books.
On Kindle, with sideloaded books, chapters are included as are the annotations you have previously removed.
Fortunately, there are many tools online that automatically remove these duplicate entries. This is because the annotations are stored in a dummy MyClippings.txt file. It isn't a database or some strange JSON file, just a simple .txt file that separates its annotations with "====" symbols.
You can even go a step further and create a simple Python yourself script to delete those duplicated entries and I wonder how would that work with the SQL database of Kobo.
One thing I would give to Kobo is that neither of the other two alternatives can include images in their highlight exports. Kindles don't include them to counter piracy, KOReader doesn't include them because it's probably one pull request away. The band-aid solution to this is to simply take a screenshot and organize the images later.
And guess which of the three handles screenshots the worst?
KOReader prompts you to choose a folder path the first time you take a screenshot. Kindle, until very recently, didn't have a separate folder for them.
Kobo, to this day, August 8th, 2026, not only requires you to add a new line to the configuration file, but still drops the screenshots into the root directory.
Keep in mind that once you add that line of code, the Power Button stops powering off your device and instead acts as the screenshot button. So, in the very understandable case that you wanted the Power Button functionality back, you either have to remove that new line of code or install NickelMenu to add a toggle between the two functions.
Genius!.
It's 1 in the morning and tried verifying the status of my database after setting up the Google and Dropbox Integration and found out my database had already been corrupted (missing books and unused pages). Up until this point I had never plugged my Kobo into my computer.
Speaking of NickelMenu, I read a lot of users describing how much better Kobo is when it comes to modding because it doesn't have to be jailbroken to install extensions. And it is true, you can pretty much create and install any piece of software on the Kobo by either repackaging it as an update and start it through a NickelMenu menu option.
And this would be one of Kobo's strongest advantages over Kindle if it weren't for the fact that Kindle's own community and extension ecosystem has always been considerably larger in comparison to Kobo's, especially in recent years following Amazon's controversial decisions and the growing movement of users trying to take back ownership 1 of their devices.
You need to jailbreak your Kindle first in order to enjoy many of the extensions found on MobileRead or GitHub, yes, it might not be available to you right now, but when you do, you get access to multiple extensions such as KOReader, a native Anki tool, a full-blown music player (KinAMP), Python 3.11, AbiWord, automatic online screensavers, a dedicated audiobook player, a Kindle Cast streamer, a Gargoyle Interactive Fiction port, and even a rather buggy Game Boy emulator.
With Kobo, a device that's already open by default, you have access to a custom menu creator (Nickelmenu), KOReader, Plato (an open-source alternative reader similar to KOReader), a few aesthetic patches through kobopatches, a hyphenation fixer, a visible clock mod while reading, and... that would probably be it.
I can understand why the community is currently more focused on Kindle. Not only is it the most popular e-reader, but Amazon's decisions has also given the community an additional rebellious motivation to modify and take greater control over their devices which would obviously strengthen the extension ecosystem around it.
But my point here isn't to say that there should be more users interested in Kobo's ecosystem. If that were my argument, I would have removed the past few paragraphs and started creating my own extensions.
My point is that people shouldn't, conveniently claim that Kobo's openness alone makes it better than Kindle. Yes, it is open, but only open to relatively few extensions. The fact that you can install KOReader without a jailbreak doesn't automatically make it better than the Kindle. Because once you jailbreak a Kindle, it becomes considerably more extensible than what the open Kobo community has ever offered.
And I know what you're probably thinking: why would I need extensibility or specialized tools for the simple task of reading books? Why would someone need Anki on an e-reader designed to do nothing more than read a couple of books?
Well, Carol, some people do like to extend their device's functionality. And if they can do that, we can also talk about how easily they can achieve that functionality in the first place. This question doesn't necessarily apply to me, because I still tend to only read books without looking for any extra functionality. But I find the presence of an active and thriving community very convincing as well.
If someone wanted to tinker with an e-ink device, if someone wanted that extra functionality, I would unfortunately have to point them towards a jailbroken Kindle.
If you even ask on Reddit about what other extensions exist for Kobo, you will very likely get unironic recommendations for Calibre plugins. They're not terrible suggestions, but they're unconsciously disingenuous because users asking that question don't expect Calibre plugins as an answer. Even today, you can get suggestions for the KoboTouchExtended plugin, which has already been superseded by a previous Calibre update.
Retrospective
Playing devil's advocate here, I very much prefer Kindle's cloud reading experience to anything KOReader or NickelMenu could achieve. Why? Because it's much more simpler and streamlined than what KOR and Nickel could ever do (Kobo's Google Integration does not sync annotations and progress).
I am well aware that I can achieve something similar by using KOReader, store the .sdr folders and pass them through a cloud service or a Calibre-Web docker instance and pray anything of it works on a random Tuesday... but I also think it's perfectly reasonable to say that not everyone wants to set up Proxmox and a Calibre-Web docker instance just to keep track of their library or reading progress.
I know Syncthing can also do the job by syncing the metadata folders, but it isn't available on iOS, and the last time I tried setting it up between an Android device and my Kindle's KOReader installation, it worked about as jankily as you might expect from an open-source project without a QA department.
I am pretty sure I could tweak it a little more to finally get it working the way I want it to. Maybe I'm missing a singular page of documentation that could have fixed all my problems. But I would sooner or later end up in the same mentality that seems to plague the power users in the community, they end up tweaking and adding new functionality instead of spending time reading their books. You would be underestimating how easily developers can create dozens of new problems while trying to solve one.
KOReader is much more powerful software for reading books than either Kindle or Kobo, and I believe I will ultimately end up using it on my Kobo for the time being despite not liking it fully. Because it's that same power that makes me wary of it, I can understand why some people stay away from it. Even when doing the simple task of reading a book, its endless extensibility, customizability, and intricate menus can quickly disorientate someone or create a lot of friction, which is something that usually happens to me.
There's also the notion that KOReader drains battery off the Kobo's significantly faster, a main developer has replied with the observation that Kobo's battery gauger isn't reliable2 , but I think it's not normal for it to drain 20% of my battery after reading for one hour and a half, and yes, with Bluethooth and Wi-Fi turned off.
I've had my fair share of issues with Kindle's cloud reader and synchronization, but despite that, I would still consider it a more much more comfier experience. I don't have many worries about Amazon "owning my books" because I would usually send books that I have already bought somewhere else. If I wanted to buy a book that was only available on Amazon and nowhere else, I would usually wait for its exclusivity contract to end and buy it from Smashwords, Bookshop, or even the publisher's official website.
And even if I was worried about my privacy, I would do the same thing I did for my previous Kindle device. I created a "fake" Amazon account in the past that was associated with a McDonald's in Dallas and a fake United States phone number and identity thanks to my MySudo subscription.
Not everyone has that patience for this but it works for me. If I wanted even more protection of my privacy, I would use the "Block Device Logging" option from the BinaryRenamer extension and call it a day.
Perhaps the biggest lesson from this switching process is understanding that having more control over a device does not necessarily mean having a better experience with it. I wanted the freedom Kobo promised, but I end up realizing it was missing a ridiculous number of things that I had taken for granted on Kindle.
And I was truly not that interested in truly owning my books, because I already had a huge library of epubs before getting my first Kindle e-reader. I think I will eventually buy another Kindle, upload my books, and get over it. I won't have to worry about sync, reading progress, or metadata. I would probably complain a little about the broken updates, but after seeing the other side of the wall with Kobo and KOReader, I don't think I would have much ground to complain about.
Would I recommend a Kobo? If you're interested in living within the KOReader ecosystem and can put up with a few days of customization and probably a lifetime of tweaking your system, then I would.
If you're someone who wants to truly tinker with their e-reader, I would recommend a jailbroken Kindle, despite the fact that it will still be relatively locked down even after the jailbreak, the true value comes with its very active community.
But if you're my coworker, or me, who is slightly interested in reading a little bit more, I would probably just recommend the latest Kindle. It would be much easier for both of us.
References
YouTube video search of people wanting to take back ownership from their Kindle. You can also own your books (and most other stuff) by not buying everything from a predatory platform that exclusively rents and not sells.↩
NiLuJe's comments on the Noticeable Battery Drain on KOReader thread, 2021↩