Orwell be proud.
TL;DR Formerly unusably-slow for backups a few years ago, the built-in Target Disk Mode in Macs has improved in latter-day Intel models (using a Thunderbolt 3 cable), now nearly 1/2 as fast as an external USB-3 drive dock.
My test: use Carbon Copy Cloner v5.1.22 to clone an internal SSD with 690GB of data. “Find & replace corrupted files” is OFF*.
Source: Macbook Pro 16″, Catalina 10.15.7
Targets: external SSD (usb 3) vs. 2020 13″ MBP (TDM, thunderbolt 3)
Result: Direct-connect two MBP using a high-speed Thunderbolt-3 cable was 5.9 GB/minute, about half as fast as backing up via USB 3 to an SSD in an external drive dock (12 GB/minute).
Keep in mind that two “trips” are required if the goal is to move the data to a new Mac via an external SSD (first from source -> external SSD, then from external SSD -> target)… so when you add the time it also takes to restore the external SSD to the target Mac (which I did not measure), the net time difference between the two approaches is less.
For my purpose, simply cloning Mac#1 to Mac #2, the direct-connect using Target Disk Mode was simpler and adequately-fast. YMMV.
|External SSD, cheap Insignia Dual Dock||USB 3||56:32|
|Internal SSD of a 2020 MBP 13″ Intel (updated to Catalina 10.15.7) in Target Disk Mode, writing to a new APFS volume**||Thunderbolt 3, direct connect, using 40G Pluggable cable plugged into Left*** ports||1:56:23|
- FWIW the time to use CCC’s “Find & replace corrupted files” feature just to confirm file integrity, with essentially no new data being written, was 3 hours using Target Disk Mode; e.g. 50% longer than the time to clone the disk to a new volume with the feature disabled.
** Because APFS makes it so easy, I now keep the “out of box” OSX in the original APFS volume when I get a new Mac (as a backup bootable volume, updated to the lastest version of Catalina), use Disk Utility to add an empty APFS volume for work, then clone my old drive to that new work volume.
*** I didn’t test it myself to quantify any different, but have seen claims that the right ports have lower data transfer speeds.
When it was time to pay, my surf instructor today asked if I use Venmo.
Venmo has IMO ridiculously high default transaction limits (thousands of dollars) and does not (according to their support rep), give users ANY way to lower them.
Which seems like some evil bullshit to me, given they offer zero recourse in the case of a transaction that goes awry.
They advertise themselves (in big letters) as “Fast, safe, social payments”.
Who in the hell makes “social payments” transactions of $2,000, and weekly total “social transactions” of $4,999?
Okay, maybe some of you do.
But I’d guess most of us wouldn’t.
If they’d let users set their own lower limits (if desired, for example transaction limit size to say $100 or $150), I would use Venmo from time to time to make the occasional payment for coffee, splitting a lunchtab, or pay my surf instructor.
But, they are IMO too-clever, too-greedy asshats.
“No way,” I replied to to my Surfing instructor. He then pulled out his little Square reader, where I have far greater protection against transaction mistakes than Venmo’s lame policy: “Venmo Support can only reverse a payment if the recipient gives their explicit permission”
Transaction limit safeguard options are always Good.
Companies that won’t let you set them, and also provide no recourse in case of mistakes, are Bad.
If more people would “just say no to Venmo” (and tell them why) maybe they’d listen.
If I’m being honest, at first I might have been a little flattered when the Amazon chat support rep suggested I “bare” with him/her. It’s not my thing, so I declined, but still… nice to be asked. (Once we hit 60, those offers are few and far between.)
Then I realized, wait, maybe it’s just a new form of multi-factor authentication. After all it’s creepy how much info big-web collects about us, and webcams get hacked everyday, so maybe they just want to take a good look to ensure it’s really me, not some other geek in a button-down shirt posing as me, before they answer my question.
Problem: Apple 27″ Thunderbolt Display not working when connected to 16″ Macbook Pro (2019) via the Apple Usb-C to Thunderbolt 2 adapter. The display remained dark, despite have just been disconnected from an older Macbook running Mojave.
First (simplest) solution: Connecting a stand-alone Thunderbolt 2 cable from the Macbook adapter to the BACK of the Display worked fine, the Display works fine.
But since I was pretty sure the display’s T2 cable is fine (even if it did not work connected directly to the Macbook) I then tried putting a Thunderbolt peripheral between the Macbook and the Display (using the Display’s built-in T2 cable). And that worked too, freeing up the T2 port on the back of the Display!
Second (better) solution: Display’s built-in T2 cble connected to a T2 peripheral (and OWC thunderbolt dual drive dock in this case) and then the peripheral’s other T2 port connected to the Macbook’s adapter.
If you use other T2 peripherals, solution 2 is superior since it frees up the extra T2 port on the back of the Display.
Apple still refusing to give product owners a way to disable Apple’s ineptly executed “smart” links is unfathomable. Its annoying on a Mac and an iPhone, but inexplicably stupid on Apple Watch.
An 8 digit product number – obviously not a US phone – gets highlighted as a phone number, tap it and it just places a call to… well, NOT to the USA.
Ditto a 5 digit zipcode, also highlighted as a phone number link to call who-knows-where.
But is the actual URL in the text message clickable? Hell no.
If you own domains, there are situations where you would certainly want to be alerted if an issue arises… for example if a domain registration renewal failed due to an old credit card number.
So it’s a great idea for a registrar like GoDaddy to offer a text message – in addition to the usual email – if something is expiring, for example.
Until the marketing assholes decide to force you to also accept marketing messages if you want important tech/admin messages.
One of the larger sites for finding and hiring Filipino virtual assistants is Onlinejobs.ph (headquartered in Utah).
Not surprisingly, the site uses Filipino workers to build and maintain their own website… artists, support staff, programmers, etc.
No doubt they save a lot of money by using Filipino developers.
But if you’re going to save money on developers, for goodness’ sake double-down on Testing. (“Testing” in a broad sense to include for example functional specifications.)
As an example, here’s an incredibly stupid bug I found today on the very popular Onlinejobs.ph site… which as I write this is adversely affecting the site’s utility for both employers and workers. (And no telling how long it’s been broken, perhaps since day 1.)
The website hosts job postings from employers, and resumes of job seekers, and offers employers a way to keyword search for job seekers, and offers job seekers a way to keyword search for job postings.
So the fundamental CORE of their service is SEARCH so employers/workers can find each other. The rest is window dressing.
But in an example of inept programming PLUS inept testing, their search is fundamentally broken in the sense that if a job post is open to “any” type of worker (full time, part time, or freelance) a job seeker searching for full-time will NOT SEE that job post. The only job seekers who will see that job post are ones who ALSO specify full time AND part time AND freelance.
(In other words, their search should “OR” the job post work type -but it stupidly does an “AND”.)
Hey, that kind of bug can happen, especially when you save money hiring cheaper and less experienced developers.
Which makes it all the more important to NOT skimp on testing methodology implemented by someone actually thinking about the test cases. It would have caught this bug before it failed untold times to correctly match workers and employers.
(Another bug on this site is the ongoing inability of their PHP programmers to figure out “escaping/unescaping” submitted text so job posts containing a percent symbol or an ampersand symbol render correctly. But those at least are obvious by proofreading unlike the search bug.)
I finally terminated my Prime membership … the last straw was the heavy-handed and arrogant non-support at Amazon.
I’ve posted 69 prior reviews.
Tried to post the following review of an outstanding new novel I purchased on Amazon, which review Amazon rejected. Their chat support claimed it was a technical error that would be fixed, but the next day I got an email saying “Amazon is unable to post the customer review because some of the details of this account indicate a relationship to sellers, publishers, or other reviewers of the product.” I asked for specifics, heard nothing back.
Heavy-handed and authoritarian.
My review of “Deep River” by Karl Marlantes:
Move over, Michener.
I can only make time to read a couple novels a year, but my interest piqued by the favorable WSJ review, I downloaded this to my Kindle before a trip and was amply rewarded.
This beautifully crafted saga is an epic view of the melting pot that was America during the tumultuous early 20th century… families compelled to leave their ancestral homeland, and how in the process of making America their new home they made America so much more than the sum of its parts.
Marlantes’ writing is a joy to read, richly nuanced characters and historical detail, punctuated with gem-like aphorisms such as “Aino’s first impression of America was elbows and noise.”