Jolla Tablet refunds | Hacker News Reader

Jolla Tablet refunds

blog.jolla.com

38 pointstype020 comments

If you still want a Jolla tablet, there is an indie project started. They had all the blueprints and designs of the original tablet device. Licensing of Sailfish OS should have been taken care of. The first step is to bring out a device with most of the same hardware as the Jolla batch. In the future there might be updates on the hardware, most probably with a 3G module.

http://talk.maemo.org/showthread.php?t=99214

Great to hear. Oh and maemo forum is still active, I remember it fondly during the years of nokia internet tablet.

Here in the US, having a 3g/wifi tablet would be a waste of money, because the two major GSM carriers are putting most of their spectrum on their respective LTE bands. T-Moible is planning on sunsetting 3G as well.

I'm willing to forego my refund, as long as Jolla will open source SailfishOS now. Community has been waiting for it like forever.

I'm an iOS developer and I mostly love iOS design and user experience but one OS that, imo, is better is MeeGo.

I used the N9 briefly for 3 months back in '12 or '13 and other than the ecosystem issue, the OS was intuitive and simple to use.

That being said, I'm slowly moving onwards to React Native projects so hopefully in the near future, my paths will re-intertwine again.

Is developing a mobile OS significantly harder than a desktop one? It's odd to me that many FOSS projects do a fine job with desktop Linux distribution yet it is so hard to have a mobile one?

Nokia spent 1B Euro on MeeGo? Did Palm have that kind of money when they did webOS?

The thing is that thanks to IBM mistakes, the desktop became kind of open, while on the mobile market there are lots of NDAs and OEM deals to take care of.

You also see this affecting the desktop with the transition to laptops, most of them being like appliances.

Desktops are standardized: you run on a i386 clone... Of course over time the supported hardware has expanded, but there are a lot of standards to work with. In particular this means that the kernel boots, X starts, and the network connection works. Once you have that can build anything.

Mobile has never really worked for FOSS because there is no standard to work with. By the time you get the kernel booting the manufacture has obsoleted that hardware and you don't boot on the next one so you have to figure out why and get it working - just as you do that the hardware is obsolete again. You still haven't got graphics on the screen, or a network connection. Graphics isn't hard and is often finished for those who bought the system before hand, but networking is generally not: mobile means a cell connection that is secret.

The funny thing is that this tablet was/is an Intel tablet. A lot of parts will probably follow laptop standards. The problems in this case were a bit diverse. I think an investor or manufacturere pulling out was the biggest reason.

It could work if the hardware vendor released all the platform-specific code as open source. Then a community could potentially keep that code up to date well enough to interface with more rapidly changing layers on top of it.

Of course, it is more work than if the hardware is standardized. But it might be doable if you don't have to reverse-engineer things.

I think FOSS had the most early success with older hardware that was relatively inexpensive and still fit for its purpose. They often made excellent developer machines for people who wanted to try a Unix-like system for the first time.

Tablets, OTOH, have been becoming obsolete within a year or so of use and don't seem to have much value after that. This might change now that the pace of change has started to stabilize and there are more tablet Chromebooks starting to make it onto the market. I wouldn't be surprised if a good FOSS solution starts to emerge once tablet-like Chromebooks start to gain momentum.

For an alternative way of thinking, maybe there already are some decent FOSS solutions for tablets. There are a lot of actively developed third party Android distributions for tablets and ChromiumOS is pretty powerful.

> Tablets, OTOH, have been becoming obsolete within a year or so of use and don't seem to have much value after that.

Huh? A persistent problem of all tablet manufacturers was that tablets change so little over the years, that people are perfectly content with continuing to use their 4-year-old iPad.

That's at the high end. The low end of the market is pretty much beset by churn. Even if people aren't replacing older models, the older models are quickly being retired so porting an OS is a moving target.

Yes, that is true for iPads, but those are really hard to unlock and are unlikely to see FOSS development.

"High-end" Android tablets from four years ago haven't held up nearly as well.

Its a good question. And as many comments hear have pointed out the difference is variance.

When you're designing a complex system which is a combination of hardware and software there are many many questions which can be answered in two multiple ways. One of the variables "how much to do in hardware" vs "how much to do in software". The answers have ramifications all over the system. In desktops all of the 'what to do in hardware' options are fixed by the "PC" spec.

As a result, all of your development effort, goes along the path of writing code, and then refactoring it if it doesn't do what you need it to do.

In a mobile system you can make choices, and making those choices impacts how you write the software. You can find problems that require you to change the hardware and all of the software (for example). As a result it is harder because you have more of the design space to work in, and its more expensive because working through that design space takes time, and time is literally money when you're paying salaries and leases.

A mobile operating system running on mostly closed hardware is going to cost a fortune in reverse engineering and debugging. I'm sorry if I keep stressing this point but we don't need another mobile OS, especially when it's that hard to make one for the above reasons. Instead we absolutely do need an open platform; give us a phone/tablet board where everything (except maybe cellular radio) is completely open and documented, and dozens OSes will follow. Somebody or some entity with the necessary power should push for truly open platforms, though I expect neither Microsoft nor Google or Apple to have any intention to do that.

The worst thing is I don't think it would be too hard to make a quite open phone, even an upgradeable phone. From the top of my head:

* Raspberry pi compute module as a base

* SIM8008 equivalent with 3g/hspa module soldered on a daughterboard, with battery charging circuit

* I2C sensor package (accel, gravity, typical AKM modules)

* lcd with usb touchscreen

You could make a nice DIY phone (I've been wanting to make something like this from an old Nokia 9210 or something like that). If you could make more powerful versions creating some boards with different SoCs respecting the CM pinout layout, you could make some nice kit to put on kickstarter, but you would also need a software stack to support it, and everything is so targeted these days that THAT would be the most difficult part. It's not hard to make hardware talk each days, it's integrating it with the software everybody else is using. What's a phone without Whatsapp / Telegram? How can you download an app without a store? People don't use forums these days to get to know new software... Sure, you could make lineageOS run on it, but then, that would differentiate much from other phones, except for the tinkerers and OSS freaks (me included), but there are not that many of us as to make that a profitable business

I agree that we need an open hardware initiative in phones, but you need a shitload of money to burn before you can make any kind of difference in the established ecosystem, and would still not be too sure who the real target would be

linux runs tons of embedded devices as well (i.e. resource constrained devices that make your phone look like a super computer).

Refunding based on random order is patently absurd. They should refund in order that orders were taken in, not randomly. That's not fair at all.

Don’t all the people who need a refund have an equal right to it?

Reply on news.ycombinator.com