Given those points, what is the issue with this estimation using bytes?
My observation stands. Do with it as you please.
Gee, thanks for nothing.
I'll go out on a limb and assume you mean, "yeah, I guess that's fine for this estimate, especially given the the limited info we have to go on, but I don't want to admit it."
What? Like, "meets minimum job requirements"?
The way you're analyzing this shows you don't have extensive subject matter expertise. Hint: think in tokens, which varies a bit by model, not bytes.
1. How many tokens are in a "100-Page prompt"? Good luck answering that from the info provided.
2. Did you look at the doc I linked?
Given those points, what is the issue with this estimation using bytes?
What kind of car do you drive? I bet there is no way to repair it if it needs modules replaced other than with windows software.
Um... ECUFlash, OpenOCD, TunerPro, TunerStudio, CANutils... You sure?
It's also not a product most people directly purchase. Most would have a repair shop fix it. I don't think "product that requires Microsoft" meant at any point in the history of any component within the product; I think it means in order to actively use it in it's normal day to day operation. The ECU in my car does not run Windows, and I don't need a Windows license to run my car. The water treatment plant may might have some Windows systems doing various things, but I don't consider water to be a product that requires Microsoft. If you do, we're not using language the same way.
If you're programming vehicle modules and your setup requires Windows 11 and said software doesn't play well with VM's and you've determined the consequences of issues are too high (vehicle modules)... why would you be ok with kludging together some unsupported hardware and jumping through hoops to hopefully get Windows 11 to continue working? Aren't the consequences of failures too high?
It's gotta be an incredibly narrow set of edge cases that:
* Can't be moved to Linux
* Can't be run via Wine or related forks
* Can't be run from within a Windows VM (or risks of doing so are too high)
* Won't continue running on Windows 10 with extended security updates
* Can run on fully supported Windows 11 machines
* Don't justify purchasing new hardware that fully supports Windows 11, and/or you can't afford such hardware
*
Maybe there's an example out there, but it sure seems like most people pulling out such excuses simply don't want to do the alternatives. "I don't want to deal with doing $X right now and would rather limp along with what I know for now and/or ignore the inevitable cause I dread what's coming," seems entirely reasonable and relatable, if a bit delusional.
Personally, I'm kicking that can down the road - turned off my Windows laptop around 8 months ago and it's just doing nothing now (was only used for Teams here and there, and a cheap chromebook has filled that role). FWIW, it's already dual booting Linux; I just don't need a laptop often.
No one has paid an upgrade fee.
So in your view, there's no scam nor upgrade fee if you simply (*checks notes*) pay them more so you can postpone the upgrade fee???
That's not "remain in support"; That's "pay to buy additional/ongoing support". And that's apparently the good alternative?!?
You might as well say, "No one has to pay an upgrade fee. It's free to install Linux." Which, though snarky, is more honest than what you're claiming.
Maybe they don't have an account themselves with access but if they commit some code that gets promoted to production and runs with account privileges that do...well bob's your uncle.
Right on. The fact that he had a process, one he named after himself, checking if his own account had been disabled, and that the disabling of his account was the trigger to do stuff, means said bit was using other credentials (not his own).
I'm not sure hiding processes would be considered basic knowledge.
At a defense contractor writing software? It should be a prereq for getting the job.
Use of RAG for the data seems appropriate, but the context window issues are very real. I found this to be very informative:
https://ancillary-proxy.atarimworker.io?url=https%3A%2F%2Fgithub.com%2FNVIDIA%2FRULE... - "RULER: What’s the Real Context Size of Your Long-Context Language Models?"
TFA doesn't give anything more specific than "100-page prompt". Very rough estimate on size:
* 100 pages * about 250-500 words per page = 25,000 - 50,000 words
* my
* so roughly 1/4 - 1/2 that, or 250K - 500K
The open source model from that list with the largest claimed context size has a 1M context size. HOWEVER, the largest *effective* length is somewhere between 128K - 256K.
Assuming these are professionals who actually spent significant time on this (as TFS claims), they'll be well aware of these issues and limits. They probably used every bit of the context window that they could feasibly use, as that sounds like it'd be awful close to the limits of the open source models.
Anyone happen to know if/what models include the output in the context window? 25 more pages may be really stretching its memory.
Aisle gets you an bit more leg room if you place them in the row.
I disagree. For me (192 cm / 6'4") sitting that skew for extended periods is a pain in the backside (literally) - butt and beyond. Not to mention all the people and catering trolleys bumping into your lower appendages. I also have broad shoulders, so impinging on seat neighbor's as well as aisle space, and getting bumped into more. I still like the aisle seat for being able to stand and walk as I please, even in the middle of the night. Not necessarily for bladder reasons, just to get limbs going.
Thankfully, domestic flights are short, and long haul expensive enough for me to not need them. The last couple of years I only flew overnighters due to work, and they always mandate the cheapest flights, and the seating is almost a skin-tight fit in the knee and lap area with the little table out (but still survivable). Maybe it's a blessing in disguise that they are struggling with finances and have cut travel to the absolute necessary (which excludes us mere underlings).
100% this
No, updates are applied effectively without any indication to the user. Most users don't even know they had an update if the app itself doesn't announce its new features / fixes.
Sounds like you're saying "Fuck your use cases do something else". To quote someone you know, "has {that} approach worked for anything other advice you have given?"
They took away features that lots of users were using with legitimate reasons, but your argument is that you want to "use the software I want to use with the features I want". Yeah... that's what we're saying too. Isn't that fucked?
Define "plenty" as a percentage of Windows Home users. Hell, define it in terms of the overall workforce.
Then tell me how much effort is justified in catering to them.
There is zero effort needed to cater to them. The support was already there. They expended effort to exclude existing uses, and the underlying code is still there - only with new license-specific limitations.
I can't define that percentage, as I can't even find a stat on the percentage of Windows 11 installs that are home versus pro. I can say that basically everyone sleeps at night, and anyone with a home license will get hit with unexpected reboots. Any of those people with stuff open that won't survive a reboot is at risk.
What number of impacted users would it take for you to consider the extra effort they went through to disable this feature to be a waste? With 1 billion+ Windows 11 installs (which includes home and pro), even a sliver of a percent is a hell of a lot of installs.
I'm sorry, that came off harsher than I wanted. I'm not changing it, but I want you to know I'm not trying to be mean.
Appreciate the note, and ditto from me. It's slashdot and we're just nitpicking obscure tech stuff. No worries.
The update issues you brought up are not common in a home setting. They are more common with the software people use at work, and then they come to us to complain about it. That just doesn't happen very often at home.
If by "not common" you mean "happens to less than 50% of users", you may not be completely wrong. However, the number of users that does impact is HUGE.
According to a quick google, there's over a BILLION devices running Windows 11. I didn't see any quick stats on home vs. pro, so take that with a grain of salt. If only 10% of those are home users, and just 1% of Windows 11 Home users have such issues, that's still over 1 MILLION impacted users.
They didn't have to add any features or registry entries or special handling; They simply had not not purposefully block the use of the existing registry keys and features for home users - features that are still there in the same codebase when a different license is applied.
AND NONE OF THAT EVEN MATTERS! The whole argument you're making is for the subset of users that aren't comfortable touching settings, let alone registry entries. For that subset, automatic app updates were ALREADY enforced. Your argument doesn't justify one iota of this change. Are you just angling for the "subtle troll" mod? WTH?
This is nonsense, what if it is not a window but a door instead, should you be suing based on a technical name?
Is there a window in the door? Then that's fine.
Window seat is seat adjacent to a wall, aisle seat is a seat adjacent to a corridor.
FFS, do you hear yourself? Who the hell defines an aisle seat as next to a corridor? BTW, corridor definitions refer to passageways in buildings from which doors lead into rooms, political corridors, or strips of land. Aisle is not a synonym (https://ancillary-proxy.atarimworker.io?url=https%3A%2F%2Fwww.merriam-webster.com%2Fthesaurus%2Fcorridor). Do I need to tell you a window is not a wall?
Honestly, I wish I had that leg room - my knees aren't what they used to be. But IME, it's more trouble than it's worth to me. Too many other people going back and forth and forcing me to move and move back. I've also been hit in the elbow by the cart while I wasn't even trying to lean out - just had the tiniest bit of my elbow over the edge of the armrest. I'd rather have the view and be able to crash out against the wall.
But it does move! -- Galileo Galilei
