An Interview With Zlatan Todoric, Open-Source Developer & Former Purism CTO

[brent]

So apparently Purism announced the end of the first batch shipping window...

https://twitter.com/Puri_sm/status/1186724741928968192

...without shipping even a single phone to a customer that is not affiliated with Purism. LOL.

[Michael]

So apparently Purism announced the end of the first batch shipping window...

https://twitter.com/Puri_sm/status/1186724741928968192

...without shipping even a single phone to a customer that is not affiliated with Purism. LOL.

If you haven't seen already, they did admit they really didn't ship to customers yet, just employees/insiders basically - https://twitter.com/vjrubens/status/1186773079629946881

[blackiwid]

So apparently Purism announced the end of the first batch shipping window...

https://twitter.com/Puri_sm/status/1186724741928968192

...without shipping even a single phone to a customer that is not affiliated with Purism. LOL.

Sure it might been another small delay for the public but in general isn't that a good news? Doesn't that mean that they NOW start with the next batch or does the delivery of the next batch start in 5 months from now?

I mean the later batch you get the better quality Product you get right? So if they would start very fast the delivery of the next batch that would be a big plus for the ones that get this one instead of the last one? They said that this was always meant as small batch, which kind of makes sense then you don't want to hand solder or bend some stuff on thousands of devices right?

Also reading their batch faq they seem to never have said people can freely choose the batch they want and then demand to get it, and it's apparently the earlier you ordered it the earlier batch you get except you opt out to prefer a later one.

So it makes sense that if that first batch was meant to be very small numbers that mostly only employers and partners get it, because they probably were the ones that ordered first, especially the employees ordered before public ordering was available or know of the order field first or the employer pays / subsidies their orders.

So after they announced that they deliver in batches they never said that they will deliver to everyone that wants batch 1 a device so I don't see this big scandal, but maybe that's just me?

Again it makes sense that they don't handcraft many in the Aspen batch:
Individually milled case
unfinished switch caps (hand crafted)

So with the Batch Birch batch it should be much better:
Shipping window: October 29th – November 26th

There you will get your devices to people that are not affiliated with them.

[danmcgrew]

From Pine64‘s "October Update: PineTime, Delays and Shipping News" ( https://www.pine64.com )

"[EDIT 18/10/2019] Since writing this blog entry I provided an update on the situation (Pinebook Pro, PinePhone and PineTime [the 'PineTime' is Pine's $20.00 'SmartWatch'] dev kits shipping) on the forum. Click here."

["Clicking here" gets you to...]

https://forum.pine64.org/showthread.php?tid=8085

...and here is ONE of Pine64's answer to a commenter, regarding 'open-ness', and the Librem device--

"Hey,
About being open: I see no other way for PINE64. In a sense this is more of a typical open-source project than a 'company', and most of it is community ran. To this end, there is no traditional divide between the device vendor and the community. You want to write for the blog or even edit the front page? - ping me. Want to edit the wiki - you can do so already ... any part or it too, just use the forum creds. Want to see something in the next products we announce? - you can talk to me and TL [the founder of Pine] any day of the week (pretty much) in the chats. Doesn't take much to convince us to do X or Y if its sound. The PinePhone is a prime example of this - it was built from the ground up with devs and community input. "

"As for the Librem5: Looks like they have a very interesting device on their hands. I wholeheartedly hope they'll succeed in achieving their goals with - the more Linux phones on the market the better for everyone. There is place for both devices ... and many more to come I hope."


Welcome to true open-ness, honesty, and lack of any rancor regarding the competition, folks.

Based on some of the self-serving comments here, it'll probably be too much for some of you to handle. I suggest you types don't read anything from Pine, or anything positive about Pine; simply keep wallowing in your well-cultivated and hard-won stupidity and pre-conceived notions.

"Do not share the truth with those who don't deserve it."--Mark Twain

"Most people will believe facts to be true, but only if those facts agree with what they already believe."--anon

[cynical]

1. Why not go for a Qualcomm chip? There was already upstream driver work, so less effort needed (and I am still not sure if the current GPU is properly hardware accelerated).
2. They want to use flatpaks, so the base system should have been an image (potentially using something like OStree) to store images like Fedora Silverblue does.

1. Wow, Qualcomm was an option? I'm not up to date on any open-source work for mobile chips, but if true that is major.
2. This is the only thing that make sense for a phone. As much as people might enjoy using apt-get, you really need images for OTA updates on a mobile device. And flatpak or equivalent for applications is a no brainer as well, but the other issue, one that Canonical was well aware of, is that you need restrictions on what applications can do through some kind of platform API (including critical services), otherwise you will inevitably get poor battery life as Android found out.

Retrofitting the existing GNU/Linux base to work on a phone is not simply a manner of making the app UIs scale to phone size, or even building the system UI for the phone. You need a proper design that starts from the ground up for mobile. Systems for handling application run times, efficient notifications, permissions for apps, etc. Even today Android is for example discouraging anyone from using SMS permissions due to abuse. They don't have the model right after all these years, so this is not an easy thing to do.

It sounds from this description like there was not adequate time spent on design upfront. This is not a good sign for the longevity of the system. Starting with UBPorts and building on top of that base would have made more sense, imo, given how much Canonical invested in doing things the right way.

[Flohack]

It sounds from this description like there was not adequate time spent on design upfront. This is not a good sign for the longevity of the system. Starting with UBPorts and building on top of that base would have made more sense, imo, given how much Canonical invested in doing things the right way.

Yeah but unfortunately Qt is not fully FOSS - that was one of the arguments they came up with when we asked what are the issues...

[You-]

1. Wow, Qualcomm was an option? I'm not up to date on any open-source work for mobile chips, but if true that is major.

They might have needed an soc that didnt have integrated telephony, but otherwise from what I see, Qualcomm seem to be playing well upstream. I think you can credit RobClark who has posted in this topic quite a bit for this, he reverse engineered driver eventually got to the state that Qualcomm engineers started to support it and AFAIK even employ(ed?) him to work on it. He has posted here to confirm that certain SoC's are well supported. The 8916 I mentioned earlier is also 4xA53 cores at 1.4ghz and its modern successor (because AFAIK they give new features/driver support for a couple or so years from release only) would have been easier to support.

There are AFAIK 4 major graphics players in the Arm world: Qualcomm/Adreno, Arm/Mali, PowerVR and to a lesser extent Vivante. The first two now have a lot of work upstream to make them behave. I dont know what the state of PowerVR is, but Vivante is one where a lot more work was required to get it into shape, especially compared to the first two.

Saying that, Purism as worked upstream and if their efforts result in a new platform haing mainstream support, we all benefit.

[AccurateVictory]

But I am pretty confident that as much people hate QT (it's ugly as hell) and love GTK/GNOME.

You are talking about a toolkit/engine, there's nothing to do with the end visual which is basically themable.
From a technology standpoint, GTK/Gnome is garbage and worst choice for mobile. You are happy on your netbook but on my core i7 3520m with 8gb, it was far to be the case. HOWEVER, i did like the visual and UX.

For my part, Canonical/Ubuntu should have gone for Qt/Plasma after Unity, let their UX/designer cleaning Plasma and it would have been a huge win but no, we have a ugly hack.

Anyway, off-topic.