OESF Portal | OESF Forum | OESF Wiki | LinuxPDA | #planetgemini chat on matrix.org | #gemini-pda chat on Freenode | #zaurus and #alarmz chat on Freenode | ELSI (coming soon) | Ibiblio


Welcome Guest ( Log In | Register )

Personal Photo
Personal Statement
Middle-aged C# developer with a history of Delphi, Pascal, OPL, Basic, 68k and 6502 machine code. While smartphones outperforms any Psion I've had, before Planet Computers, none had a reasonable keyboard. Currently using a Cosmo as my only phone, though I have a spare phone, just in case.
Personal Info
Daniel W
Age Unknown
Birthday Unknown
No Information
Joined: 22-May 18
Profile Views: 6,334*
Last Seen: Today, 04:13 AM
Local Time: Jul 15 2020, 09:39 AM
239 posts (0 per day)
Contact Information
AIM No Information
Yahoo No Information
ICQ No Information
MSN No Information
Contact Private
* Profile views updated each hour

Daniel W


My Content
18 Apr 2020
On the Indiegogo Astro campaign page, the size of the device is said to be 164(W) x 76.6(D) x 15(H)mm.

A Cosmo is (officially) 171(W) x 79.3(D) x 17.3(H)mm, and my ruler seems to agree (when not busy sticking to the magnets in the device). The depth of the Cosmo keyboard alone is pretty much 70mm, so that could perhaps fit in a 76.6mm deep Astro, but as the Cosmo keyboard in itself is 164mm wide, it reasonably can't fit inside a 164mm wide device, lest it had absolutely no side bezels at all, which all the images of the Astro clearly shows it has.

So, maybe they shrunk the keyboard a tad? I doubt that, since the new screen is about 15mm wider (a 6.53" 20:9 screen is about 151.25 x 68.06mm) than on the Cosmo (a 5.99" 18:9 screen is about 136.08 x 68.04mm) and they still need to fit the speakers somewhere.

I'd rather guess the Astro keyboard itself, excluding bezels, will be 164mm wide, while the total width of the phone will be akin to that of the Cosmo and Gemini. Whether it's a typo (maybe the correct width is, say, 174mm) or whether they just put the wrong measurement (the width of just the keyboard - not the entire device) there, or if it is, somehow, correct, remains to be seen. What do you think?
25 Mar 2020
Attached File  CoDi49.jpg ( 191.47K ) Number of downloads: 14
Attached File  CoDiKeepass.jpg ( 205.26K ) Number of downloads: 7

Well, the title says it, and, as mentioned, it's a really cheap scope, so the image quality isn't that good. Yet, it reasonably illustrates the subpixel layout, which seems to be RGB with every other line (running vertical in this image - it was hard enough to get any image at all) pushed over 1.5 subpixels.

One "pixel" might consist of a little "RGB triangle", but that's just me guessing. I don't really know anything about how the display is driven. If anyone else here have anything to add, I'd be interested.

For some reason the blue subpixels on every other line seems to be brighter (to the point of appearing white in this image). The kind of stripey pattern (running horizontally in this image) seems to affect everything shown on the CoDi. I'm not sure if that's an artifact of my particular unit or whether it's universal. Does anyone else here see this (very fine) vertical stripey pattern on their CoDi?

Edit: Added a second example.
5 Jan 2020
Glancing this Gemini related post about some outgoing IP connections that turned out belonging to the firmware updater, I found the web site of Ash Wolf (Ninji here at OESF), upon which these two blog articles:
picks apart the Cosmo Over-The-Air firmware updater, and finds, well, questionable content.

A firmware updater, reasonably, must have basically every permission, so we're kind of forced to trust whichever firmware distributor Planet Computers chooses. While I do trust Planet Computers not to be malevolent, they seem, to me, somewhat clueless at times, and, it seems, they've picked a firmware distributor whose other business, apparently, is to, via their own updater, distribute malware. Ouch.

Maybe they're only doing that as a paid service, say, on behalf of dirt cheap phone makers, who might want to make up for their low prices by exploiting their customers in any profitable way they can come up with. I'm quite certain Planet Computers isn't involved in or, as it seems, were even aware of, any such capabilities.

Yet, the way this is implemented on the Cosmo, it seems ANY app can silently get ANY Android permission, by knowing how to ask one of the updater interfaces. While nobody might specifically target such an uncommon device type as the Cosmo, probing for that interface would, to me, seem like something any competent malware author would do, in case their code happens to be on any phone where this interface is available.

As far as I understand, that can't happen, lest I'd install a malware-laden app first, but as those, according to media, once in a while, does make it onto Google Play, no matter how reasonable I'm trying to be, this feels a bit too crazy for comfort. I'm at a bit of a loss right now. After waiting over a year for my Cosmo, intending to use it as my only phone, I suddenly don't know if I could, at all, trust this device, once it arrives. Thoughts, anyone?
11 Dec 2019
As this is about the Airmail by Planet Computers app, I was a bit torn as to where to post this. As I expect to use it a LOT more, once my Cosmo comes, I put it here.

Anyway... Airmail displays HTML e-mail just fine, so I just supposed it would compose them reasonably well too, but I've found NO such support at all. This threw me off earlier today, when I needed to e-mail a step-by-step instruction with some images in it, maybe bold text in the subheadings, and red text for a warning. On the PC, I routinely do such things in Mozilla Thunderbird. There, I need to enable the HTML support to use it when composing e-mail. Is there a similar setting in Airmail, or is the Planet Computers e-mail client, supposedly specially geared for their devices, supposedly specially geared for mobile creators, really just capable of composing bare bones basic black text?

If so, there will be some e-mails I just cannot send with it, or reply to. If I am to insert snippets of source code into an e-mail, those snippets, but not the rest of the e-mail, needs to use a mono spaced (and, usually, slightly smaller) font. If I'm going to suggest changes to a text, I need strikethrough, or color, or... something to make a clear difference between what I want to remove, and what I want to insert. I need ??????? (hrm... it seems IP.Board can't handle Unicode symbols too well) and super/subscript for basic math notation, or footnotes, lest ?(X2+Y2) is to degrade into SQR(X^2+Y^2). You get the idea.

Can any of this be done in Airmail, or can the community, pretty please, suggest a Gemini/Cosmo-compatible e-mail client that does?

It doesn't have to be free, but it can't just be a web mail, and it must work with regular e-mail servers, so some special Gmail-only app, or similar, won't do. I don't need anything too fancy, it's e-mail after all, but, well, I roughly need to be able to send this forum post (and Unicode symbols) by e-mail, without loosing anything. Yes, sure, I can write and attach a Word™ document, but only if I know the recipient has Word too, which they may or may not - especially if they are on a mobile device too - or their organization might be wary of such attachments. In a pinch, I could send a PDF, which almost everyone can read, but those are deliberately designed to NOT be editable. So, yeah, there are painful workarounds, but - just to be clear - I'm not asking for those. I'm asking for somewhat capable e-mail, as I'm used to in the non-mobile world.
13 Jul 2019
With Update #20 out and, including today, 19 days left of July 2019, perhaps it's time to start guessing when:
A: The first user Cosmo production run will start.
B: The first user Cosmos leaves the factory.
C: The first user posts on OESF that they have received their Cosmo(s).
D: Most users who backed their Cosmos during the crowdfunding month have received their units.

A: As consumer electronics certification must be fairly streamlined, testing only HAS to find issues that cannot be fixed via software updates, and they're currently waiting for more of the material they've already got for PR1,
I'm guessing Monday, July 29. Correct answer: Tuesday, September 3 (so my guess was off by 36 days).

B: For the Gemini, the first user mass production commenced around Saturday, January 20, 2018 (Update #44). Three weeks later, on Saturday February 10, it was complete (Update #48). I seem to recall it took longer before units actually were shipped from the factory, but as I hope things will be a bit smoother this time,
I'm guessing Monday, August 19. Correct answer: Friday, October 11 (so my guess was off by 53 days).

C: The earliest post here from a user who received their Gemini, says it arrived on Thursday, March 1, 2018, which is close to six weeks after the first production run was complete, but as I think that was due to some shipment issues,
I'm guessing Monday, September 9. Correct answer: Friday, November 8 (so my guess was off by 60 days).

D: I got my Gemini three weeks later, on Wednesday, March 21, 2018. While I had to wait a bit, as, apparently they assembled "other" keyboard layouts (such as Swedish) last, I wasn't the last early funder to get mine, but, again, as I hope things will be a tad quicker this time,
I'm guessing Monday, September 30. Correct answer: Thursday, January 23 (so my guess was off by 115 days).

There. Yes, I've simply guessed "three more weeks" each time, as I think it's more or less reasonable. Naturally, I HOPE things will be faster, but, well, shipping stuff from China isn't typically fast. Now, it's you turn to guess, should you want to. Then we can compare, both to the mass production schedule we're supposed to get "soon", and, later, to actual reality.
Last Visitors

19 Jun 2020 - 7:21

2 May 2020 - 23:21

27 Mar 2020 - 3:18

26 Mar 2020 - 6:09

21 Mar 2020 - 13:42

Other users have left no comments for Daniel W.

There are no friends to display.
RSS Lo-Fi Version Time is now: 14th July 2020 - 11:39 PM