kimchi

joined 1 year ago
[–] kimchi@lemmy.world 1 points 3 days ago* (last edited 3 days ago)

Dati and Bournazel have an alliance which may provide Dati with most of that 11.3%:

https://franceinenglish.com/p/rachida-dati-and-pierre-yves-bournazel-unite-in-strategic-alliance-for-2026-paris-municipal-elections

Knafo's 10.4% is obviously never going to Gregoire, bringing that combined vote to 47.2% (minus, say, 25% leakage, yielding ~42%).

Gregoire could gain Chickirou's 11.7%, minus leakage, yielding ~47% (of previous voters).

Then there will be voters who didn't participate in the 1st round. It's close, but I'd still give the edge to Gregoire.

[–] kimchi@lemmy.world 2 points 3 days ago (1 children)

This is brilliant: tire wear (loss of X grams of rubber) scales with distance traveled, vehicle weight, speed and aggressive driving.

but the supply chain is very hard to lock-down, without tire serial numbers.

[–] kimchi@lemmy.world 2 points 6 days ago (1 children)

Day-to-day, it's very nice. Charging at CCS chargers is very fast (though charging at Tesla is slower than some, 98kw). If you're not in a warm climate, be sure you get one with battery preconditioning (in '22, that meant getting an AWD model).

The GT (and I think GT-Line?) has a low ceiling due to the sunroof... we got the Wind+options due to tall family members.

The interior and exterior 120v outlets are really nice: I powered a freezer, fridge, window A/C unit and internet from it during an all-day power outage.

[–] kimchi@lemmy.world 4 points 6 days ago (2 children)

AFAIK, all EV6's, IQ5s etc have the ICCU issue, which appears to be under-spec MOSFETs . One tally on a forum I'm on estimates 2%-10% of owners have experienced failure. However, the 10yr/100k-mile electric warranty covers it.

There is a new part number last month for the ICCU (from old 36400 1XFA0 to new 36400 1XFA0A), so there's some hope that it may be sorted. Maybe your dealer can verify that a new EV6 has the new ICCU part number (I've also heard that it can be read via CarScanner).

We are at 80k miles, so hoping either ours pops within 20k, or else there is a recall or class-action.

[–] kimchi@lemmy.world 1 points 2 weeks ago* (last edited 2 weeks ago)

The only Fast Charging most EV owners do is on road trips. The rest is more like plugging your cell phone in while you sleep. So the relevant comparison is: how long do you usually stop for a bio-break & snack+checkout. I wish I could get the family in and out a convenience store as fast as the EV6 charges (though it's much slower than Blade2's high-speed charge).

Of course, most petrol users fuel-up weekly in the USA, so the petrol car is starting each road trip at a disadvantage. If you fuel-up with petrol for 4 minutes, 4x/month, and road-trip 1x/month, then the petrol car starts each road trip 16 minutes behind.

[–] kimchi@lemmy.world 1 points 2 weeks ago

Sounds like it's got a built-in battery about the size of 26 77kwh car batteries, and is maybe fed 24x7 by a 100 amp, 480v 3-phase charger.

This sounds like the "FreeWire Boost" chargers in US... except 12.5x as much batteries.

[–] kimchi@lemmy.world 13 points 3 weeks ago (1 children)

Be sure to identify as "an avid driver... but sometimes I want to bike on Central, and it's just not safe!" ;)

26
submitted 3 weeks ago* (last edited 3 weeks ago) by kimchi@lemmy.world to c/privacy@lemmy.ml
 

Using GrapheneOS, my main profile has a handful of apps from PlayStore(Aurora): 1password, ProtonVPN, ProtonMail, etc.

I think I read somewhere that, for an app to appear in PlayStore, it must be compiled with linked libraries that implement check-ins with Google infrastructure... or something like that.

Obviously I'd expect apps like 1password and Proton to be "less evil," but am curious whether everything from playstore leaks telemetry, or if it's just "up to the developer".

(in my case, I don't have Google services or apps in the main profile at all)

[–] kimchi@lemmy.world 1 points 3 weeks ago

Same here: I pay contactless with a Garmin vivoactive3. It cost $30 on Craigslist, runs for ~3 days without charging. My thought was that Garmin has less "ecosystem" and is less likely to be monetizing purchase history, vs. Samsung... but I don't have any inside info.

I have the GarminConnect app in my "Google Crapola" profile, and for a few months I even deleted the app, and tap-to-pay on the watch kept working.

[–] kimchi@lemmy.world 2 points 3 weeks ago

I tend to buy <$200usd phones and use for 4 years (so: ~$50/year). I'm on a Pixel6A now, and plan to upgrade to an 8A at EOL in summer 2027. An unlocked 8A in "good condition" is $255 now on Swappa, and would get updates until summer 2031, so that meets the $50/year metric.

I'll be interested in GrapheneOS's new partner, but I suspect only their brand new phone will work, so... I'll consider it in 2031 :)

[–] kimchi@lemmy.world 2 points 1 month ago* (last edited 1 month ago) (1 children)

The way I read it is:

  • if you never plug-in overnight, and the vehicle is big, and you drive aggressively, you get 34mpg (believable)
  • but if you plug-in a small car every night, and you get 75% of your miles electric, and you drive like a grandma, then you get 223mpg (believable)

Sadly, it sounds like Porsche drivers may fall into the first category and Toyota drivers in the second. And there are enough Porches to skew the MPG of the whole PHEV class.

(it's also possible that Porsche/VW/Audi just make PHEVs that score well on gov't tests but poorly in the real world, though I'd lean towards the drivers. But the article title really implies that all PHEVs get shockingly bad mileage)

[–] kimchi@lemmy.world 24 points 1 month ago* (last edited 1 month ago) (1 children)

The article is horribly unclear: it seems to say that PHEVs are no good, but "the main reason for the higher-than-stated fuel usage was ...that the PHEVs use two different modes, the electric engine and the combustion engine". Well, so do non-plugin hybrids. I doubt they're saying that plug-in hybrids are worse than non-plugin, but you might guess that from the title.

The article states that Porsche PHEVs used 7 liters per 100 miles (33.6mpg), but Kia/Toyota/Ford/Renault used "85% less" (1.05L/100k or 223mpg... maybe about right if driven 75% from plug-in energy).

Porsche mentioned "different usage patterns". I can buy that a typical Prius owner is plugging-in every night, filling low-rolling-resistance tires to 54psi and driving like grandma, and a typical Porsche owner... isn't. If you want apples-to-apples, then compare a gas Corolla vs a Prius vs a Plug-in Prius, where the cars are from the same city/suburb, and similar owners (e.g.: no ubers, no regional sales reps).

This "study" is evaluating real-world use of one class of vehicles, and not other vehicle types; then using the dismal ways some people drive to imply that this particular class of vehicles is the problem.

[–] kimchi@lemmy.world 1 points 1 month ago

sorry, I think I was confusing bank transfers with credit card transactions.

 

I was wondering if the $4K(USA) used EV tax credit was still getting applied at USA auto dealers. It sounds like the credit still applies until September 30th, but I haven't heard if the IRS/dealer website and database are still working.

1
submitted 10 months ago* (last edited 10 months ago) by kimchi@lemmy.world to c/networking@sh.itjust.works
 

Has anybody heard if the upcoming 47-day maximum on TLS cert lifetime will apply to Enterprise wifi auth using private PKI (especally on IOS and Android)?

We have a campus CA that signs the TLS cert used by RADIUS when students connect to wifi using personal devices. Freshman need to accept the cert once (hopefully after checking the fingerprint), then usually one more time before graduation. Every 47 days would be difficult.

 

I have banking apps in a separate User profile. I was wondering if this was preferable (or worse) than putting those apps in Private Space.

Anybody have a "Separate User vs Private Space" comparison?

1
submitted 1 year ago* (last edited 1 year ago) by kimchi@lemmy.world to c/linux@lemmy.world
 

I'd like to use the box.com cli from a linux server without X installed. However, during "$ box login -n me@mydomain", boxcli tries to popup a web page with the message Opening browser for OAuth authentication. Please click Grant access to Box to continue.

I've tried exporting BROWSER=lynx , and have tested that "xdg-open http://eff.org/" works; but I am not seeing lynx open while running "box login". Do I need a 2nd terminal window, and then set DISPLAY so the "popup" browser opens lynx there?

Wondering if anyone has managed to setup box.com CLI on a headless, X-less server.

 

I'm blowing the cobwebs out of my mom's 1986 Ward's (Happy Sewing Co) machine. I have been watching videos of setting timing:

adjust timing until the hook passes through the scarf...

...and how to set the needle bar:

adjust needle bar height until the hook passes through the scarf...

(I'm paraphrasing)

It sounds like you could take a perfect machine, then lower the needle bar 1mm, then compensate by delaying the hook 30 degrees, and you'd have the hook passing through the scarf at the correct spot... yet it would be all wrong.

Is there a way to set needle-bar height independent of the hook timing?

Like, obviously the needle needs to rise a few millimeters to make the slack thread form into a loop behind the scarf, ready to be caught by the hook. Is that amount of rise kinda-sorta consistent across machines from a given era?

view more: next ›