If you could swap your smartphone for another device, what would that be?
You're strongly encouraged to explain your response.
#Smartphone #DumbPhone #FeaturePhone #DeathOfTelephony #Comms
@neil No, but this is something I've been actively researching for ... an embarrassingly long time (months) ... with low satisfaction. But as this is on my current to-do list, this toot doesn't count as procrastination 😺
I have a perfectly serviceable early-2000s flip phone which I'd love to reactivate ... but it won't talk to any current-generation mobile cellular networks.
I've looked at current low-feature phones, with interest in the Light Phone and Punkt MP02. Both ... have drawbacks, though I'll likely end up with one or the other.
The MP02 lacks comprehensive protocol coverage, and hence is limited to a subset of local mobile providers, unfortunately the ones I'd most prefer not giving my business. (I'd prefer not giving any of them my business, but that seems unlikely).
The MP02 does offer integrated Signal messaging, which is a strong plus.
The Light Phone is spendy but is also compatible with more carriers, which ... will probably win me. There are a couple of revs on the initial design, though I'd favour e-ink (original) over OLED (LPIII).
What both the MP02 and LP offer is tethering, which means your phone becomes your no-WiFi backup for Internet access, enabling you to use a more-featured, non-telco device for comms.
Which suggests another path:
Palmtop PCs (e.g., the Psion 3) are no longer of this world, which is a shame. But smaller laptops, e.g., Framework's 12" (8.45 x 287.00 x 213.88mm, 1.3 kg), are effectively a far-more-capable larger tablet (I've been carrying a 13.3" Onyx BOOX for the past four years), without all the handicapping and crappification of Android. With a Bluetooth earpiece that becomes its own vox comms device, and you have full access to any other comms protocol through your Linux distro of choice, with a full keyboard. The Framework 12 is convertible (360° hinge) with stylus support. It doesn't have an e-ink screen module (yet?), but ... I can live with that. For anything other than on-the-run use, it's likely superior to any smartphone or tablet.
The other issue is vox comms management. I'd prefer not being directly on the PSTN network, so what I'd really like is that my "phone" accepts (and mostly makes) calls to only one other point: a VOIP relay which intercepts all incoming calls and redirects outgoing. Incoming traffic is then limited to only those numbers I allow, or which pass some screening mechanism. And my mobile telco provider gets a Very Boring profile of whom it is I'm in contact with. Whether or not this ideal VOIP service / capability exists I'm not sure, but that's what I'm looking into.
And yes, it would be possible to ditch the phone entirely for a pure-play modem, though the ability to make calls directly in an emergency has its arguments.
@riley Those adjustments are still themselves based on statistical methods.
The underlying situation burrows a bit further into communications options.
Household penetration of landline phones hit ~90% by the late 1960s (https://transportgeography.org/contents/chapter1/the-setting-of-global-transportation-systems/household-telecommunications-united-states/). Not only was that a fast and cheap way to reach most of the public, but the namespace was densely filled and could readily be randomly walked, and on a local basis by using area codes and local exchange numbers (the first three digits of a traditional seven-digit phone number).
That is, with a "wardialler" dialing random numbers, and excluding those which were not in service or institutional (business, gov't, schools, etc.), it was possible to quickly reach a useful representative sample of the public (~300 to 1,000 people, typically), which is all you need to get a reasonably representative view WITHOUT any regard for how large the total population is. Ironically, it's often easier to get a represntative national sample than it is for a small region such as a city or county. You could select given area codes to address a state (or portion of one), and exchanges typically identified specific towns or neighbourhoods of larger cities.
Almost all problems arise from sampling bias, which is what your weighted samples and other factors try to address.
Household landline penetration is now at or below 20% in some US states, though others have higher values than I'd have though, NY topping the list at 52.4%. Mobile devices and many landlines now block unknown (or specifically blacklisted) callers. Robocalling is an absolute plague.
All of which means that there's currently no universally utilised telecommunications means of running surveys. The options are to statistically adjust known methods, to do in-person or postal-mail-based contacts, or find other alternative methods of contact. The #DeathOfTelephony problem I discuss from time to time is having its impacts on US politics and polling, among other areas.
Another solution to phone spam occurs:
The first time a call is made to a new number, the telco intercepts that call and requires a (random, 3--4 digit) code be entered.
Humans would be able to enter that code readily. Robocalls ... not so much.
But wait, there's more!
The connecting telco knows where that call comes from in terms of the connecting network. It's either a local origin (on the same network), or remote. Which means that the telco can monitor the acceptance rate for calls coming through.
The telco can, and should, also solicit feedback from the call recipient of whether or not the call was spam.
Networks which exceed some reasonable threshold for spam call volume would be denied access to the recipient's network for some period of time. I'm a fan of exponential backoff. So, for the first phone spam threshold reached, access is denied for 1 minute. If further spam is received within 4x that interval, it's doubled (2 minutes, 8 minute window), and again 4/16, 8/32, 16/64, 32/128, 64/256, 128/512; 256, 512, 1024. With 11 spams, the denial is > 1 day. 13x is a week, 15x is a year.
Originating networks would be strongly encouraged to clean up their acts.
How would you suggest fighting robocalls and phone spam?
Friends in the US are reporting that over 45 of the past 50 calls they've received are robocalls, telemarketers, hang-ups, misleadingly-identified, or otherwise smell strongly of fraud.
STIR/SHAKEN was rolled out two years ago and has quite obviously failed. See: https://www.theverge.com/2021/6/30/22557539/t-mobile-verizon-carriers-fcc-stir-shaken-certification-deadline-spam-calls
I'm looking for systemic solutions here, not personal mitigations. "I've stopped using the phone and am now only communicating by Ansible" may suit you very well, but it doesn't address the billions of people who do have directly-addressable voice coms as mobile or landline service.
Examples of systemic solutions:
Think regulation, making global changes to software or hardware, changing switching and call handling systems, or market-based interventions, such as:
Things of that nature.
NOT "I downloaded this app and use it on my pocket spy device."
Boosts very much welcomed.
@lordbowlich Yeah, I've been seeing the #DeathOfTelephony for some time, and most especially an end to general direct-dial telephony.
I'm inclined toward a mobile-network-dongle model where that is effectively a firewalled connectivity service which I can discard at will, whilst my own comms take place through that, but abstracted from it, and I can swap out different bits of network kit.
Somewhere there's a hub which does track where I'm at, sends contact notifications (if I'm interested in / permit those), and otherwise manages interactions.
Getting there from here will be bumpy, and we'll lose many of the affordances of the present system, though we seem to be well on our way to losing those already.
@mathew Split the logic.
Sell a cheap, disposable, baseband unit which tethers to the main compute platform.
Added bonuses (bonii?): segregation of mobile and core identities. Swap tethers as desired.
PSTN's days are numbered as it is.
@alexandra Important, but not the sole metric.
What users are present also matters, and if the high-value / valued-contributor / advertiser-friendly users are leaving, then both other users and advertisers rapidly lose interest.
Remember, Facebook was Once Literally Harvard. It's no longer that.
North America + Western Europe are about 1 billion of the richest people on the planet, and amount for the lion's share of FB revenues. Stalled growth in either market (happeend in NA in 2018) or decline hits hard.
Worse (from a business perspective) if it's affluent / telegenic members. Not all nodes are created equal, and loss of high-value members hits disproportionately hard.
This is the other end of the network-effects lever arm, and has an asolutely devastating effect on network value and business viability. Wireline telephony has hit this point for the past decade, and I suspect all direct-dial teleophony isn't far from this point (see my #DeathOfTelephony toots).
#DeathOfTelephony poll: Interesting results, and considerably more optimistic than the last time I'd run this.
I'd really like to hear from those selecting "32 years +" (there should be about 14 of you), and either 4 or 8 years (likely about 6 in total) what informs or guides your choices.
How many years until General Direct Dialed telephony dies?
Discuss.
(Bounds are "greater than previous, less than/equal to value", excepting the final option.)
Boosts welcomed.
Doc Edward Morbius ⭕
